Insulin resistance (IR) is the pathophysiological hallmark of type 2 diabetes and obesity. Although several biological pathways have been proposed to be behind IR, there is still a great deal of uncertainty about the mechanisms by which cells and organisms become insulin resistant. The etiology of IR involves intricate interactions between genes and the environment, interactions mediated by epigenetic mechanisms. DNA methylation is a major epigenetic modification that is performed by DNA methyltransferases (Dnmts). Altered patterns of DNA methylation associate with a variety of metabolic perturbations, yet the cause-and-effect relationship remains poorly understood. Our recent studies have demonstrated that IR associates with many locus-specific changes in epigenetic modification. While we were searching for differentially expressed epigenetic modifiers, we discovered that adipose expression of Dnmts was elevated in mouse models of IR and that the increased expression was largely reversed by the insulin sensitizer Rosiglitazone. We have shown that Dnmt3a, in particular among the Dmnt family, is both necessary and sufficient to mediate IR in cultured adipocytes. Consistent with our in vitro results, our primary in vivo studies found that adipose-specific Dnmt3a-knockout mice had improved whole-body insulin sensitivity and glucose tolerance. Furthermore, our gene profiling studies identified Fgf21 as a key target gene repressed by Dnmt3a, with concordant changes in DNA methylation at its promoter. Based on these findings, we hypothesize that adipose Dnmt3a mediates IR by regulating key metabolic genes through altering site-specific DNA methylation at critical cis-regulatory regions. To test our central hypothesis, we will pursue the following aims. Aim 1 is to ascertain the role of Dnmt3a in cell-autonomous IR and to delineate the epistatic relationship of Dnmt3a and PPAR?. Aim 2 is to investigate the full impact of adipose-specific Dnmt3a depletion on whole-body metabolism and to elucidate the underlying mechanisms of the improved insulin sensitivity, and Aim 3 is to elucidate the underlying molecular and epigenetic basis through which Dnmt3a mediates IR using both focused and unbiased approaches. Overall, this research will establish the role of adipose Dnmt3a in the pathophysiology of IR and elucidate the underlying molecular and epigenetic mechanisms. It may also provide novel biomarkers of IR and reveal therapeutic target(s) for IR.